Version bump 0.6.0 -> 0.7.0

Fixed: remove astra-toolbox 2.3.0 support from package metadata

Add NumPy 2 compatibility (#37)

Added: New AutogradOperator for easier autograd support (#27)

Torch: Improved Monai interoperability of to_autograd #25

Update README.md to use new pytorch and cuda packages

Fixed wrong dimensions in learned_pd example code (pull request #24)

Fix positional args passed to kwarg-only function

Fix error message in AssertionError in OperatorFunction

Version bump 0.5.1 -> 0.6.0

torch: slightly changed test_autograd_shape

readme: references and links to ts_algorithms were added

torch: Improved interface of operators made with to_autograd

Version bump 0.5.0 -> 0.5.1 related to #16, also added unit test

Fix buffer datatype initialization for non 32-bit data